[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#280206: apache: Apache wont start, FD_SETSIZE set too low in 1.3.31-7?



No, the kernel is the same as before the upgrade. 

I guess I could have just hit the roof of some other limit, I do add vhosts every day. It just seemed unlikely to me that it stopped working just then.

I don't really care if this bug(?) was there before or not though. Is there anything we can increase without messing up something else that can fix this? I have this in my /etc/init.d/apache: 

[snip]
case "$1" in
  start)
    should_start
    test_config
    echo -n "Starting web server: $NAME"
        ulimit -n 1000000
    $ENV start-stop-daemon --start --pidfile $PIDFILE --exec $DAEMON > /dev/null
    ;;
[snip]

Notice the ulimit -n command. But this doesn't help (I think it did before though, but im not sure). 

I havent tried building the source package myself, mostly because i dont know how to do that ;)


----- Original Message ----- 
From: "Fabio Massimo Di Nitto" <fabbione@fabbione.net>
To: "Peter Bredlöv" <peter@scorpionshops.com>; <280206@bugs.debian.org>
Cc: <control@bugs.debian.org>
Sent: Monday, November 08, 2004 8:43 AM
Subject: Re: Bug#280206: apache: Apache wont start, FD_SETSIZE set too low in 1.3.31-7?


> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> tag 280206 moreinfo
> thanks
> 
> Peter Bredlöv wrote:
> | Package: apache
> | Version: 1.3.31-7
> | Severity: important
> |
> | I recently upgraded from 1.3.31-6 to 1.3.31-7. After this, apache wont start
> | any more unless I disable the log files I have for my around 2000 virtual
> | hosts.
> |
> | This is a sample row that shows up in my main error log when I try to start
> | with logging enabled:
> |
> | [Tue Nov  2 00:36:40 2004] [warn] make_sock: problem listening on port 80,
> | filedescriptor (1069) larger than FD_SETSIZE (1024) found, you probably need
> | to rebuild Apache with a larger FD_SETSIZE
> |
> | Although things are working ok without logging, it would be nice to be able
> | to get logging to work again even for this relatively large number of vhosts.
> | Could this be fixed?
> |
> | Regards,
> | Peter Bredlöv
> 
> Did you also change the kernel on this machine? there have been no changes at all to apache on that
> front and i suspect that either you increased the number of vhosts or the kernel is limiting
> something else.
> 
> Fabio
> 
> - --
> Self-Service law:
> The last available dish of the food you have decided to eat, will be
> inevitably taken from the person in front of you.
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.4 (GNU/Linux)
> Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
> 
> iD8DBQFBjyOghCzbekR3nhgRAuC1AJ9zr2/uwe98OQmwcUQ0Nc6dSAKBQQCeKtsw
> L/olb7H8/ZpR7O5nBAfQTIQ=
> =3P/g
> -----END PGP SIGNATURE-----



Reply to: